cheapest slavable cdp

Nov 23, 2006 at 6:50 PM Post #16 of 23
Quote:

Originally Posted by Garbz /img/forum/go_quote.gif
I do not know the differences between word and superclock, but I am more interested in this jitter thing.

The PLL I know is not perfect. Jitter can leak through the PLL from the source when adjusting a VCXO to synchronise the clocks.



Word clock is generally quite low in frequency, in the kHz. It is actually used to define the Left and Right channel multiplexing. Superclocks are generally used for the MCLK or master clock, which ranges from 11-24MHz depending on the sample rate. MCLK is usually either 128 or 256 times the word clock or L/RCLK frequency. The bit-clock is 64 times the word clock because there are 32 bits in the frame for the left and 32 bits in the frame for the right channel.

In most digital audio systems, the PLL must track the incoming data rate, whether it is from a computer or from a transport. If it doesn't, then there will eventually be underruns or overruns. The PLL has a reference clock, which may be a Superclock that sets the fundamental frequency of the PLL. However, the PLL modifies this frequency, maybe using both analog and digital techniques to match the incoming data rate. If the data is jittery, then the PLL is wandering always trying to match it. This is what creates jitter in the PLL.

The only way to completely isolate the data jitter from the clock is to buffer the data and burst the data at high-speed, or to control the speed of the source from the DAC.

Bursting at high-speed
With computers, this can be done by using USB asynchronous protocol which bursts large chunks of data at high-speed. With a transport, this can be done by reading the data at high-speed like that done on portable "shock-proof" CD players.

Controlling the source from the DAC
There are a few systems that try to do this and achieve some level of effectiveness. Lavry DAC's are one of them. The Empirical Audio Pace-Car is another. It usually requires a large buffer and control going back to the source of the data stream.

Steve N.
Empirical Audio
Manufacturer
 
Nov 23, 2006 at 8:19 PM Post #17 of 23
did u customer use the transporters analogue out or use it as a transport with another dac?
 
Nov 23, 2006 at 9:58 PM Post #18 of 23
Quote:

Originally Posted by Konig /img/forum/go_quote.gif
did u customer use the transporters analogue out or use it as a transport with another dac?



He used the analog outs on both SB and transporter, not the best choice.
 
Nov 23, 2006 at 11:02 PM Post #19 of 23
Quote:

Originally Posted by audioengr /img/forum/go_quote.gif
Controlling the source from the DAC
There are a few systems that try to do this and achieve some level of effectiveness. Lavry DAC's are one of them.



I'm not sure which Larvry DAC you are talking about but none to my knowledge control the source. Rather they all have a very innovative method of eliminating the jitter which can propegate as noise through the PLL to the VCXO's control circuit by buffering the data and monitoring the buffer. If it starts getting full it clicks up the speed slightly, a feedback system rather than being built synchronous via a PLL.

Now if only USB drivers were easier to write
rolleyes.gif
 
Nov 23, 2006 at 11:16 PM Post #20 of 23
Quote:

Originally Posted by audioengr /img/forum/go_quote.gif
He used the analog outs on both SB and transporter, not the best choice.


exactly. i wanted to use the transporters word clock input function to slave it to my lessloss.

I will get your pace car if i can get my hands on a zanden 5000 dac.
 
Nov 24, 2006 at 5:05 AM Post #21 of 23
Quote:

Originally Posted by Garbz /img/forum/go_quote.gif
I'm not sure which Larvry DAC you are talking about but none to my knowledge control the source. Rather they all have a very innovative method of eliminating the jitter which can propegate as noise through the PLL to the VCXO's control circuit by buffering the data and monitoring the buffer. If it starts getting full it clicks up the speed slightly, a feedback system rather than being built synchronous via a PLL.

Now if only USB drivers were easier to write
rolleyes.gif




Yes, I'm aware of the Lavry concept. It is a control system with feedback to the clock source. I recently sold an Off-Ramp Turbo 2 to a Lavry DA10 owner. He reported that the lower jitter of the Off-Ramp Turbo 2 did improve the clarity a bit. It is probably one of the best DAC's on the market for rejecting jitter.

I am with you on the USB drivers. I may spend megabux next year to get an asynchronous one written for me. Still waiting on the bid.

Steve N.
Empirical Audio
 
Nov 24, 2006 at 5:11 AM Post #22 of 23
Quote:

Originally Posted by Konig /img/forum/go_quote.gif
exactly. i wanted to use the transporters word clock input function to slave it to my lessloss.

I will get your pace car if i can get my hands on a zanden 5000 dac.




Did you read the review on the Zanden in Stereophile? Not exactly glowing.... I dont believe it will pass 24/96 either.

The Pace-Car will work with my Off-Ramp I2S and a computer. It is I2S input and I2S output. Will work great with my I2S input Benchmark DAC-1 as well as the P-3A, Spoiler, Audiologic and Northstar DAC's. Just like having the master clock (Superclock4) at the D/A chip and clocking it directly.

Steve N.
Empirical Audio
 
Nov 25, 2006 at 5:44 AM Post #23 of 23
Quote:

Originally Posted by audioengr /img/forum/go_quote.gif
FIY - I just spoke with a customer that had SB3 and bought a Transporter hoping to get something audiophile quality. He said the sound was not great, so he is sending it back.
....
Steve N.
Empirical Audio
Manufacturer



I guess I'll have to take this one with a bushel of salt.
wink.gif
 

Users who are viewing this thread

Back
Top